TWI324762B - Improved audio coding systems and methods using spectral component coupling and spectral component regeneration - Google Patents

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an audio encoding and decoding apparatus and apparatus for transmission, recording, and playback of audio signals. In particular, the tree informs the input or records the audio information given that the required information is reduced while maintaining the perceived quality of the reproduced output signal at the desired level. L· «tr ^ Background of the Invention Many communication systems often exceed the available capacity in the face of information transmission and recording capacity requirements. As a result, there are considerable related issues in the field of broadcasting and recording to reduce the amount of information required to transmit or record audio signals that are intended to perceive people without degrading their perceived quality. There are also related issues to change the perceived quality of the signal at the required material or storage capacity. Traditional methods for reducing the need for information capacity include the selection or transmission of only selected portions of the input signal, while the rest are ignored. Conventional techniques, such as general perceptual coding, convert the original audio signal into a spectral component or a frequency sub-band signal, and those redundant or unrelated signal portions can be more easily identified and ignored. If the signal portion can be reproduced from other parts of the signal, the part k is considered redundant. If the signal portion is perceptually meaningless or inaudible, then the portion of the signal is considered to be irrelevant. The perceptual decoder can regenerate the missing redundant portion from the encoded signal, but it cannot produce any missing irrelevant information that is not redundant. However, the fragmentation of the Syrian customs is that there is no effect on the yarn, so the loss of the mosquito-free is acceptable. The credit coding technique only picks up those recorded or perceptually unrelated, and the 5 tongues of the u °ρ injury are sensiblely transparent. The perception that transparent technology cannot achieve the need to substantially reduce information capacity ... requires that perceptually non-transparent technology discards are not redundant and are perceptually relevant additional signal components. It necessarily leads to fidelity/deterioration of the perceived perception of the transmitted or recorded signal. Preferably, the perceptually non-transparent technique simply discards portions of the signal that are considered to be of minimal importance. A coding technique known as Lightweight, which is often considered a perceptually non-transparent technology, can be used to reduce information capacity requirements. According to the = technique, the frequency components in the input audio signals of two or more groups are grouped by the group Q to form a channel signal having a composite representation of the spectral components, and the combined channel signals are combined to form a composite representation. Ancillary information representing the spectrum packet of the wound is also generated. The encoded signal containing the coupled alpha channel G 5 Tiger and affiliate information is transmitted or recorded for decoding by the intended receiver. The receiver generates a de-synchronization signal by generating a copy of the coupled channel signal, which is an inaccurate copy of the original input signal and uses the auxiliary information to adjust the spectral component scale in the replicated signal' thus the spectrum of the original input signal The packet is roughly restored. A general coupling technique for a two-channel stereo system combines the frequency components of the left and right channel signals to form a composite high-frequency component of a single signal to generate a spectrum of high-frequency components representing the original left and right channel signals. Attached information of I. An example of a transition technique is illustrated in the Digital Television System 1324462 Committee (ATSC) Standard Document A/52, "Digital Audio Compression (AC-3)," which is incorporated herein by reference in its entirety. The information capacity requirements of the channel signal should be chosen to optimize the trade-off between the two competing needs. If the information capacity requirement of the affiliate information 5 is set too high, then the coupled channel will be forced to be low, accurate. The spectral components are transmitted at a level. The lower accuracy level of the spectral components of the coupled channels may result in audible levels of encoded noise or quantized noise being injected into the decoupled signal. Conversely, if the channel is lighted The information capacity requirement of the signal is set too high, and the ancillary information will be forced to transmit the spectral packet at a low frequency. The lower level of the spectral packet detail may result in the spectral level and shape of each decoupled signal. The audible difference in the general. Generally, if the ancillary information transmission has the frequency sub-band spectral level of the critical band bandwidth of the system of human hearing, then Good Xuanxin can be achieved. It should be noted that the decoupled signal may maintain the spectral level of the original spectral components of the original input signal, but they generally do not maintain the phase of the original spectral components. If the coupling is limited to high frequency spectral components. The loss of this phase information is imperceptible, because the human auditory system is relatively insensitive to phase changes, especially at high frequencies. The auxiliary information generated by traditional integration techniques is generally spectrum vibration. The measurement of s 〇. As a result, the decoder in the general system calculates the different scale factor based on the measured spectral amplitude, <energy measurement. These calculations generally need to calculate the square root of the sum of the squares obtained from the ancillary information. Need for considerable computing resources. Sometimes called "high-frequency reproduction," (HFR) coding technology is - knowing 7 general 'if the auxiliary information transmission has the same frequency as the frequency band of the human auditory system A good compromise between the spectral levels of the bands can be achieved. As with the coupling techniques discussed above, the ancillary information generated using traditional HFR techniques is typically a measure of spectral amplitude. As a result, the decoder in a typical system calculates the scaling factor based on the measurement of the energy derived from the spectral amplitude. These calculations generally require the calculation of the square root of the sum of the values obtained from the ancillary information, which requires considerable computational resources. Traditional systems use coupling techniques or HFR techniques' but do not use both at the same time. In many applications, coupling technology can result in less signal degradation than HFR technology, but HFR technology can achieve a significant reduction in information capacity requirements. HFR technology can be advantageously used in multi-channel and single-channel applications; however, coupling technology does not provide any of the advantages of a single channel application. SUMMARY OF THE INVENTION 3 SUMMARY OF THE INVENTION It is an object of the present invention to provide a technical improvement over the signal processing of the coupling and HFR techniques in an audio gamma system. In accordance with one aspect of the present invention, a method for composing one or more sets of input audio signals includes the steps of: obtaining one or more sets of baseband signals and one or more sets of residual signals from the input audio signal Where the spectral component of the baseband signal is in the first set of frequency subbands and the spectral components in the residual signal are in the second set of frequency subbands not represented by the baseband signal; obtained in the second set during decoding The measurement of the energy of the spectral components of the signal generated in one or more of the frequency sub-bands; the energy measurement of the spectral components of the residual signal; the spectral components in the residual signal and the synthesized signal are used to obtain the energy amount The scale factor is calculated by measuring the square root and the ratio; and the scale information representing the scale factor and the signal information representing the spectral components in the baseband signal are combined into the encoded signal. According to another aspect of the present invention, a method for decoding an encoded signal representing one or more sets of input sounds sfl # is included in the steps of: obtaining scale information and signal information from the encoded signal, wherein the scale The information representative uses the scale factor calculated by obtaining the square root and ratio of the spectral component energy measurement and the sigma signal information represents the frequency component of one or more sets of baseband signals' and wherein the spectral components in the baseband signal represent a spectral component of the input audio signal in a set of frequency sub-bands; generating a correlated synthesized signal for the baseband signal, the composite signal having spectral components in a second set of frequency sub-bands not represented by the respective baseband signals, wherein The spectral components of the correlated synthesized signal are scaled by multiplication or division according to one or more sets of scale factors; and one or more sets of output audio signals are generated. The output audio signal represents an input audio signal and is derived from Spectral components in the baseband signal and associated composite signals are generated. According to another aspect of the present invention, a method for encoding a plurality of input audio signals includes the steps of: inputting an audio signal to obtain a plurality of baseband signals, a plurality of residual signals, and a set of coupled channels. a signal, wherein a spectral component of the baseband signal represents a spectral component of the input audio signal in the first set of frequency subbands and a residual signal spectral component represents a rounded tone in a second set of frequency subbands not represented by the baseband signal The gas spectrum component of the 1324462 signal, and the spectral components of the light-weighted channel signal represent a composite of the spectral components of two or more sets of input audio signals in the third set of frequency sub-bands, resulting in an energy measurement of the residual spectral components. And two or more sets of input audio signals represented by the consumed channel signals; and the group 5 combines the scale information derived from the energy measurements and the signals representing the spectral components of the baseband signal and the stalked channel signal The information becomes a set of encoded signals. According to a further aspect of the present invention, a method for decoding an encoded signal representative of a plurality of input audio signals includes the steps of: obtaining control information and signal information from 10 encoded signals, wherein control information is derived from spectral component energy quantities The measurement is derived and the signal information represents a frequency component of a plurality of baseband signals and a set of coupled channel signals. The spectral components in the baseband signal represent spectral components of the input audio signal in the first set of frequency subbands and are coupled to the channel. The spectral component of the signal represents a composite of spectral components in a third set of frequency sub-bands of two or more sets of input 15 signals; a correlated synthesized signal is generated for the baseband signal, the synthesized signal having a baseband not The signal is represented by a spectral component of a second set of frequency sub-bands, wherein the spectral components of the associated synthesized signal are scaled according to control information; and the two generated from the coupled channel signal are represented by the coupled channel signal Group or more groups of input audio signals, the decomposed signal Having a spectral component of a third set of frequency sub-bands calibrated according to control information; and generating a plurality of round-trip audio signals representative of the input audio signal from the spectral components of the baseband signal and the associated synthesized signal, wherein The output audio signals of the two or more sets of audio signals are also generated from the spectral components of the respective decoupled signals. Other aspects of the present invention include apparatus having processing circuitry for performing various encoding and decoding methods, and transmitting A medium that utilizes a device-executed instruction program (which causes the device to perform various encoding and decoding methods), and a medium that conveys encoded information representing 5 input audio signals generated using various encoding methods. The various features and preferred embodiments of the present invention are understood by reference to the description The content and graphics discussed below are merely examples and should not be construed as limiting the scope of the invention. BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 is an exploded block diagram of an apparatus for encoding an audio signal for subsequent decoding by means of a high frequency reproduction device. Figure 2 is an exploded block diagram of a device for decoding an encoded audio signal using high frequency reproduction. Figure 3 is a block diagram of a device for dividing an audio signal into a frequency sub-band signal having a range adapted to reflect one or more of the characteristics of the audio signal. Figure 4 is an exploded block diagram of an apparatus for synthesizing 20 audio signals from frequency sub-band signals having an adapted range. – Figures 5 and 6 are exploded block diagrams of the device that are coupled to encode the audio signal for later decoding by means of a device that uses high frequency reproduction and decoupling. Figure 7 is an exploded block diagram of a device that uses high frequency reproduction and de-flashing to decode the encoded audio 12 1324762 signal. A uses the "group second analysis chopper library to provide additional frequency collisions for energy metering." The block diagram of the device that encodes the audio signal into the injury is shown in Figure 9. Figure 9 is a production发明 各种 各种 实施 实施 实施 实施 实施 实施 实施 实施 实施 。 。 [Embodiment] Mode for Carrying Out the Invention A. Overview 10 15 The present invention relates to an audio editing system and method for inputting a recording of an audio signal and encoding only a portion of a county I audio baseband portion to reduce encoding (4) information capacity requirements And the generated-synthesized signal sequentially depletes the encoded money to replace the current lost material. The New Coded Money contains the reduced poem decoding process, "Synthetic Scale Information", so the synthesized signal maintains the residual portion of the original input audio signal to a certain level of spectral level. ~

20 This coding technique is referred to herein as High Frequency Reproduction (HFR) because it is expected that in many productions the residual signal will contain higher frequency spectral components. However, in principle, this technique is not limited to the synthesis of high frequency spectral components only. The baseband signal may contain some or all of the higher frequency spectral components, or may include spectral components interspersed in the frequency subband of the total bandwidth of the input signal. Encoder

Figure 1 shows a set of audio encoders that receive a set of input audio signals and produce a set of encoded signals representative of the input audio signal. The analysis filter bank 10 receives input audio signals from channel 9, and provides a representative 13 Frequency subband information of the spectral components of the audio signal. Information representative of the spectral component of the baseband signal is generated along channel 12 and information representative of the residual signal spectral component is generated along channel 11. The baseband signal spectral components represent the spectral content of the input audio signal in one or more sets of sub-bands of the first set of frequency sub-bands, which are represented by signal information transmitted in the encoded signal. In a preferred implementation, the first set of frequency sub-bands is a lower frequency sub-band. The residual signal spectral components represent the spectral content of the input audio signal in the group or groups of sub-bands of the second set of frequency sub-bands that are not represented by the baseband signal and are transmitted without the encoded signal. In the production, the combination of the first and second frequency subbands constitutes the entire bandwidth of the input audio signal. The energy calculator 31 calculates one or more sets of spectral energy of one or more sets of frequency sub-bands in the residual signal. In a preferred implementation, the spectral components received from channel 11 are placed in a frequency sub-band having the same bandwidth as the critical frequency band of the human auditory system and energy calculator 31 provides measurements of these frequency sub-band energies. . The synthesis mode 21 represents a signal synthesis process that will occur in a decoding process that is used to decode the encoded signal generated along channel 51. The synthesis mode 21 may implement its synthesis process itself or may perform other processes that may estimate the spectral energy of the synthesized signal without actually performing a synthesis process. The energy calculator 32 receives the output of the synthesis mode 21 and calculates the measurement of one or more sets of spectral energy of the signal to be synthesized. In a preferred fabrication, the spectral components of the synthesized signal are placed in a frequency sub-band having a bandwidth that is co-located in the critical frequency band of the human auditory system and the energy calculator 32 provides an energy measurement of one of these frequency sub-bands. 1324762 in Figure 1 and in the presentations of Figures 5, 6 and 8 'showing the connection between the analytical filter and the wave mode and the synthesis mode', the built-in synthesis mode at least partially reacts to the baseband signal; This connection is optional. Some of the production of the synthesis mode will be discussed below. Some of these productions operate without regard to the baseband 5 signal.尺度 The scale factor calculator 40 receives one or more sets of energy measurements from each of the two sets of energy calculators and calculates a scale factor' as described in more detail below. Scale information representing the scale factor being calculated is transmitted along channel 41. The formatter 50 receives the scale information from the channel 41 and receives information representative of the spectral components of the 10 baseband signal from the channel 12. This information is combined into a set of encoded signals that are transmitted along the channel 51 for transmission or for recording. The encoded signal may be transmitted using a baseband or a frequency modulated communication channel that spans the spectrum from ultrasound to ultraviolet light' or may be recorded on media substantially using any recording technology, including magnetic tape, magnetic 15 Card or disc, optical card or optical disc, and detectable logo on, for example, paper media. In a preferred production, the spectral components of the baseband signal are encoded using a perceptual coding process that discards any redundant or unrelated portions to reduce the information capacity requirement. For the present invention, these encoding processes are not necessary.疋 2. Decoder Figure 2 shows a set of audio decoders that receive the decoded code of the representative-group audio (4) and produce (4) audio signal decoding 0. The signal is received from channel 59 and is encoded from the signal and _^ 15 1324762 information and signal information. The scale information represents a scale factor and the signal information represents a baseband signal spectral component having a spectral component of one or more sets of subbands in the first set of frequency subbands. The signal synthesizing means 23 performs a synthesizing process to generate a signal having a spectral component in one or more sets of sub-bands in the second set of frequency sub-bands representing the residual signal spectrum which is propagated without using the encoded signal. Figures 2 and 7 show the connection between the formatter 60 and the signal synthesizing member 23 whose proposed signal synthesis is at least partially reflected to the baseband signal; however, this connection is optional. The production of some signal synthesis will be discussed below. 10 Some of these productions operate without a baseband signal. Signal scale component 70 derives a scale factor from the scale information received from channel 61. The scale factor is used to adjust the scale of the spectral components of the composite signal produced by the signal synthesizing member 23. The synthesis filter bank 80 receives the composite signal of the adjusted scale from channel 71, receives the frequency spectrum component' of the baseband signal from channel 62 and reactively produces the output audio signal of the decoded representation of the original input audio signal along channel 89. Although the output signal is not identical to the original input audio signal, it is expected that the output signal is perceived to be distinguishable from the input audio signal or at least sensiblely acceptable and acceptable for the given application. 2. In a preferred production, the signal information represents the spectral components of a set of encoded baseband signals that must be decoded using a decoding process that is reversed by the encoding process used in the compiler. . As described above, these processes are not necessary for the present invention. 3. Filter Library 16 The analysis and synthesis filter banks can be included in any practical way including a wide range of digital filter techniques, block conversions, and wavelet transforms as needed. Production. In an audio coding system having a set of encoders and a set of decoders, respectively, as shown in the figure, the analysis filter bank 10 is fabricated using a modified discrete cosine transform (MDCT) and the synthesis filter bank 80 is utilized. A modified inverse discrete cosine transform was produced, and the modified inverse discrete cosine transform was described in the May 1987 International Conference on Acoustic Waves, Speech, and Signals, published by Pnncenetal et al. Sub-band/conversion coding of the benefit cluster design,, pp. 2161-64. In principle, no specific filter bank production is important. The analysis chopper library uses block conversion to divide the block or interval of an input signal into one. The group is constructed by representing the conversion coefficients of the spectral content of the signal interval. The cluster of one or more sets of adjacent conversion coefficients represents the spectral content in a particular frequency sub-band having a bandwidth equal to the number of private group coefficients. Filter type (for example, polyphase filter) instead of block conversion, and the resulting analysis filter library splits an input signal into - group sub-band signals. Each sub-band signal is a time-based representation of the spectral content of the input signal in a particular frequency sub-band. Preferably, the sub-band signals are substantially reduced, so each sub-band signal has the same amount The bandwidth of the number of samples in the sub-band signal of the time unit interval. The following discussion relates in particular to the fabrication of block conversion using the time domain 肖 frequency division (TDAC) conversion as in the above-mentioned time domain. In this discussion, the name " The spectral component "is related to the conversion factor and the name "frequency sub-band," and "sub-band signal" is the group associated with one or more sets of adjacent conversion coefficients. The principles of the present invention can be applied to other fabrication formats, but, therefore, the names "frequency sub-band," and "sub-band signals, are also signals related to portions of the spectral content of the overall bandwidth of the signal" and are generally known The name "spectral component" is a sample or element related to the sub-band signal. B. Scale Factor In an encoding system using TDAC conversion, for example, the conversion coefficient X(k) represents the spectral component of the original input audio signal x(t). The conversion coefficients are divided into a different set representing the baseband signal and the residual signal. The conversion coefficient Y(k) of the synthesized signal is generated when a synthesis processing program, for example, the decoding process described below is used. 1. Computation In a preferred fabrication, the encoding process provides scale information that conveys the scale factor from which the spectral energy of the residual signal is measured for the square root of the ratio of the spectral energy measurements of the synthesized signal. For the residual signal and the spectrum of the synthesized "is number can be set; the £ test can be calculated from the following expression. E{k) = x\k) (la) ES{k) = Y2{k) ( Lb) where X(k) = conversion coefficient k in the residual signal; E(k) = energy measurement of the spectral component X(k); Y(k) = conversion coefficient k in the synthesized signal; and ES(k ) = energy measurement of the spectral component Y(k). The information capacity requirement of the ancillary information measured according to each spectral component can be too high for most applications; therefore, the scale factor can be calculated from the spectrum component material or the sub-band energy measurement according to the following expression: 1324472 : m2 E(m)= ^x2{k) Λ-ml (2a) ES{m)= m2 =lr2(k) (2b) where E(m)=energy measurement for the frequency sub-band m of the residual signal And 5 ES(m) = energy measurement for the frequency sub-band m of the synthesized signal.

The sum limit values ml and m2 specify the lowest and highest frequency spectral components in the sub-band m. In preferred fabrication, the frequency sub-band has a bandwidth that is equal to the critical frequency band of the human auditory system. The sum limit value can also use a set of flags, such as A: e {μ}, denoted by 10, where {Μ} represents a collection of all spectral components that are included in the energy calculation. This flag is used for the remainder of the description for the reasons explained below. Using this flag, expressions 2a and 2b can be written as shown in expressions 2c and 2d, respectively:

£(/n)= yx2(k) (2c) k^M] 15 ES{m)= yY2{k) (2d) where {M is the set of all spectral components in the frequency band m. The scale factor SF(m) for the sub-band m can be calculated from any of the following expressions: (3a) 20 sF(m)- SS) (3b) 19 But the calculation according to the first set of expressions is usually more Effective. 2. Representation of Scale Factor Preferably, the encoding process provides scale information to the encoded signal, which conveys the calculated scale factor in a pattern that requires a lower information capacity than the scale factor itself. A variety of methods can be used to reduce the information capacity requirements of the size information. One method indicates that each scale factor is itself a number of scaled scales with a correlation scale value. One way this can be done is to indicate that each scale factor is a number of floating points, where the number of the scales is the number of scales adjusted and the associated index represents the scale value. The accuracy of the number of artifacts or adjusted scales can be used to convey a scale factor with sufficient accuracy. The range in which the index or scale value is allowed can be selected to provide a sufficient range of scale factors. The processing sequence for generating scale information may also allow two or more sets of net moving point pseudo-numbers or adjusted scales to share a common set of indices or scale values. Another approach reduces the need for information capacity by normalizing the scale factors associated with some of the base values or normalized values. The base value can be pre-expressed on the encoding and decoding process of the scale information or it can be adaptively determined. For example, the scale factor for all frequency sub-bands of the set of audio signals may be normalized with respect to the largest scale factor of the audio signal__ interval or they may be normalized with respect to values selected from the set of missing values. The base value - some indications contain scale information, so the decoding process can reverse the effect of normalization. If the scale factor can be expressed as a value ranging from zero to one, then the processing of encoding and decoding scale information can be convenient in many productions. This range can be ensured if the scale factor is normalized relative to some base values equal to or greater than all possible scale factors. Additionally, if some unexpected or rare event causes a scale factor to exceed this value, the scale factor can be normalized with respect to some of the base values greater than any scale factor that is reasonably expected and set equal to one. . If the base value is limited to the power of two, the processing procedure for normalizing and inverse normalizing the scale factor can be effectively performed using a binary integer arithmetic function or a binary bit operation. More than one of these methods can be used together. For example, a scaled message can contain a floating point representation of a normalized scale factor. C. Signal Synthesis The synthesized signal can be generated in a variety of ways. 1. Frequency Conversion A technique produces a spectral component Y(k) of a synthesized signal by linearly converting the spectral composition X(k) of the baseband signal. This conversion can be expressed as: y(j)=m (4) where the difference G-k) is the number of frequency conversions of the kth group of spectral components. When the spectral components of the mth sub-band are converted into the frequency sub-band p, the 'encoding process can calculate the frequency from the energy content of the spectral components in the m-th frequency sub-band according to the expression of equation (5) The scale factor of the sub-band P: (5) 1324762

w ίΐ^ω £5(ρ) where {Ρ}=the set of all spectral components in the frequency subband ρ;

{Μ} = set of spectral components in the mth frequency subband of the converted. The set {M} need not contain all of the spectral 5 components of the mth frequency subband and some of the spectral components of the mth frequency subband may be represented in the set more than the work order. This is because the frequency conversion process may not convert some of the spectral components of the mth frequency subband and may convert more spectral components in the mth frequency subband more than once each time in different numbers. Either or both of these cases will occur when the frequency sub-band ρ does not have the same number of components of the frequency spectrum 10 as the m-th frequency sub-band.

The following example shows a situation in which some of the spectral components of the sub-band m are omitted and other spectral components are represented more than once. The frequency range of the mth frequency subband is from 200 Hz to 3.5 kHz and the frequency subband ρ has a frequency ranging from 10 kHz to 14 kHz. The spectral components from 500 Hz to 3.5 kHz 15 are converted from 10 kHz to 13 kHz, wherein the number of conversions of each spectral component is 9.5 kHz, and the spectral components from 500 Hz to 1.5 kHz are converted into a range of 13 kHz to 14 kHz, wherein The number of conversions of each spectral component is 12.5 kHz ' A signal is synthesized in the frequency sub-band ρ. The set {M} in this example will not contain any spectral components from 200 Hz to 500 Hz, but will contain spectral components from 1.5 kHz to 3.5 kHz and will contain twice the spectral components from 500 Hz to 1.5 kHz. The HFR application description described above can be adapted to a set of coding systems to account for other considerations of the perceived product f of the signal. One consideration is that the characteristics of the converted spectral components need to be modified to ensure that the coherent phase is maintained in the red flag. In the preferred embodiment of the invention, the number of frequency conversions is limited so that the converted components maintain a coherent phase without further modification. For production, for example using TDAC conversion, this can be achieved by ensuring that the number of conversions is even. Another consideration is the noise-like or tonal characteristics of the voice SfU§. In many cases, the higher frequency portion of the audio signal is more like noise than the lower frequency portion. If the low frequency baseband signal is more similar to the tone and the high frequency 10 residual signal is more similar to the noise, then the frequency conversion will produce a higher frequency synthesized signal that is more tonal than the original residual signal. Changes in the high frequency portion of the signal can result in deterioration of the audible sound, but the audibility of the deterioration can be reduced or avoided using the synthesis techniques described below, which use frequency conversion and noise generation to maintain the high frequency portion. Noise-like features. I5 In other cases, when the lower frequency and the higher frequency part of k are similar to the tone, 'because the converted spectral components do not maintain the syndrome structure of the original residual signal, the frequency conversion can still result in I heard the sound waves deteriorated. This can be seen that the effects of sound wave degradation can be reduced or avoided by limiting the lowest frequency of the residual signal synthesized by the frequency conversion. The lowest frequency used for HFR applications should be no less than 5 kHz. 2. Noise Generation A second technique that can be used to generate a synthesized signal is to synthesize a noise-like signal, for example, by generating a sequence of pseudo-random numbers to represent time-domain signals. This particular technique has a sequence of signals that must be used to analyze the filter bank to obtain the spectrum of the resulting signal. Alternatively, a noise-like signal can be generated by using a pseudo-random digital generator to directly generate spectral components. Either one of them can be decomposed and expressed by the formula (6): Y{j) = N(j) (6) where Ν0 = the spectral component j of the noise-like signal. However, using either method, the encoding process synthesizes the noise-like signal. The extra computational resources required to generate this signal increase the complexity and cost of the programming process. 3. Conversion and noise

A third technique for signal synthesis is to combine a set of basebands for the frequency conversion of the band and the spectral components of the synthesized noise-like signals. In a preferred embodiment, the relative portion of the converted signal and the noise-like signal is based on the transmitted noise-mixing control information in the encoded signal, as illustrated by the application. This technique can be expressed by the expression (7): Y(i) = a -X(k) + bN(j) (7) where a = the mixed parameter of the converted spectral components: and b = the mixture of noise-like spectral components parameter. In a production, the mixing parameter b is calculated by taking the square root of a set of spectral monotonic measurements (SFM) which is a logarithm of the ratio of the geometric mean of the spectral component values to the arithmetic mean. Scaled and limited to vary from zero to one. For this particular production, b=i represents a set of noise-like signals. Preferably, the mixing parameter a is derived from b as follows

24 1324762 5 10 15 Surface representation: a = where C is a constant. "In a preferred production, the constant e in Equation 8 is equal to - and the noise-like apostrophe is generated such that it has a mean value between zeros and has a zero mean value and is statistically equivalent to the combined converted spectral components. The energy measurement of the energy measurement. The synthesis process can be used to analyze the spectral components of the noise-like signal. It can be used as shown above. It does not represent the converted spectral components of Equation 7t and is mixed with h. This is synthesized «Medium frequency (four) p The energy can be calculated from the following representations: Spider = Male (9) = In another production, the mixing parameters represent the specified frequency function or we specifically transmit the frequency function 3(1) and the noisy characteristics of the noisy, audio signal. How to change with frequency: In the production of the other two, the rate: the line is calculated according to the noise measurement of each = the hybrid parameter is provided by the hybrid parameter β ^ is calculated by the energy measurement of the synthesized signal Encoding and de-sequencing are achieved. The calculation of the _ component containing the noise-like signal is not the same as 'for the calculation of these energy calculations only, the encoding processing += must use additional computing resources to synthesize Miscellaneous _(four) Any other purpose of the program is synthesized and is required for the above-mentioned preferred production of the encoding process. It is necessary to obtain the spectral component energy of the synthesized signal of the expression 7 + without Synthetic Miscellaneous:: Exhibition 'Because of the spectrum component of the synthesized signal, the frequency of the sub-band energy is statistically (8) No. 25 20 1324762 is the spectrum energy of the noise-free signal. The encoding processing can be based only on the Converting the spectral components to calculate a set of energy measurements. The energy measurements calculated in this way will, on average, be an accurate measure of the actual energy. As a result, the encoding process can be based on expression 5 only from the base band. The frequency factor of the frequency band 5 is measured by the energy of the sub-band m to calculate the scaling factor for the frequency sub-band p. In another fabrication, the spectral energy measurement is transmitted using the encoded signal instead of the scaling factor. In production, a noise-like signal is generated, so its spectral components have a set of averages equal to zero and a set of 10 equals 1 variance, and the converted spectral components are modulated. Scale, so their variance is 1. The spectral energy of the synthesized signal obtained using the combined components as shown in Expression 7 is, on average, equal to the constant c. The decoding process can adjust the scale of the synthesized signal to The energy measurement has the same original residual signal. If the constant c is not equal to 1, the adjustment scale processing 15 program should also consider this constant. D. Facets are required for one of the perceived signal qualities in the decoded signal. The reduction in the information requirements of the encoded signal can be achieved by using the coupling 20 technique in an encoding system that produces encoded signals representing two or more sets of audio signal channels. 1. Encoder Figures 5 and 6 show An audio encoder that receives the channels of the two sets of input audio signals from channels 9a and 9b and produces a set of encoded signals representative of the two sets of input audio signal channels along channel 51. The analysis filter banks 10a and 10b, 26 energy calculators 31a, 31b, and 32b, synthesis patterns 21a and 21b, scale factor calculators 4A and 4B, and the format and features of the formatter 50 are substantially Same as the components of the single channel encoder shown in Figure 1 above. a) Common Features The encoders shown in Figures 5 and 6 are similar. The characteristics of the two groups produced together before the differences are explained will be discussed. Referring to Figures 5 and 6, the analysis filter banks i〇a and i〇b generate spectral components along channels 13a and 13b, respectively, which represent the respective one or more sets of sub-bands in the third set of frequency sub-bands. The spectral component of the input audio signal. In a preferred implementation, the third set of frequency sub-bands is one or more sets of intermediate frequency sub-bands that are above the low-frequency sub-bands of the first set of frequency sub-bands and are at high frequencies of the second set of frequency sub-bands Below the sub-band. The energy calculators 35a and 35b each calculate one or more sets of spectral energy measurements in one or more sets of frequency sub-bands. Preferably, these frequency sub-bands have a bandwidth equal to the critical frequency band of the human auditory system and the energy calculators 35a and 35b provide energy measurements for each of these frequency sub-bands. The consumable 26 produces a set of coupled channel signals along channel 27 having a composite set of spectral components representative of the spectral components received from channels 13a and 13b. This composite representation can be formed in a variety of ways. For example, each spectral component in the composite representation can be calculated using the sum or average of the corresponding spectral component values received from channels 13 & and jjb. The energy calculator 37 measures one or more sets of spectral energy measurements in one or more sets of frequency subbands of the coupled channel signals. In a preferred production, these frequency sub-bands have a bandwidth equal to the critical frequency band of the human auditory system and the energy meter 厶卄J 厶卄/ 37 37 provides energy measurements for each of these frequency sub-bands. The scale factor counters 44 each receive one or more sets of energy measurements from the energy calculators, 35b, and 37 and calculate the scale factor as explained above. The scale information representing the 5 scale factors of the respective input audio signals represented in the coupled channel signals are transmitted along the channels 45a and 45b, respectively. This scale information can be compiled as described above. In a preferred production, the scaling factor for each input channel signal in each frequency sub-band is calculated using any of the expressions expressed as follows:

SFM--(10)) 1〇 ring=black _ where SFi(m)=the scale factor of the frequency sub-band m of the signal channel i;

Ei(m) = energy measurement of the frequency sub-band m of the input signal channel i; and EC(m) = energy measurement of the frequency sub-band m of the coupled channel. The formatter 50 receives the scale information from the channels 41a, 41b, 45a, and 45b, 15 receives information representative of the spectral components of the baseband signal from the channels 12a and 12b, and receives information representative of the spectral components of the coupled channel signals from the channel 27. This information is combined into a set of encoded signals for transmission or recording as explained above. The encoders shown in Figures 5 and 6 and the decoder shown in Figure 7 are dual 20 channel devices; however, various aspects of the present invention can be applied to a larger number of channel coding systems. For convenience of illustration and presentation, the descriptions and graphics are only made with reference to two sets of channels. 28 1324762 b) Different characteristics The spectral components of the light channel "is number can be used in the hfr solution. In this production, the encoder can provide control information in the encoded signal for use in the decoding process to generate a synthesized signal from the channel being signaled. This control information can be generated in a number of ways. Figure 5 shows a method. In accordance with this production, the synthesis mode 213 is responsive to the baseband spectral components received from the channel 12a and is responsive to the spectral components received from the channel 13a coupled by the coupler 26. The synthesis mode 21a, the associated energy calculators 31a and 32a, and the scale factor calculator 10 40a are calculated in a manner similar to the calculations discussed above. Scale information representing these scale factors is transmitted to the formatter 50 along the channel 41a. The formatter also receives scale information from channel 41b, which represents a scale factor that is calculated in a manner similar to the spectral components of channels 12b and 13b. In an alternative fabrication of the encoder shown in Figure 5, the synthesis mode 2ia has no 15 operation with respect to one or both sets of spectral components from channels 12a and 13a, and synthesis mode 21b is not related to one of channels 12b and 13b or Operate with both spectral components, as discussed above. In another fabrication, the scaling factor for the HFR of the coupled channel signal and/or the baseband signal is not calculated. Instead, one of the sets of spectral energy measurements is transmitted to formatter 50 and included in the encoded signal, rather than a corresponding set of scale factors. This production increases the computational complexity of the decoding process because the decoding process must compute at least some of the scaling factors; however, it does reduce the computational complexity of the encoding process. 29 1324762 The second chart shows another way to generate control information. According to this production, the scale adjustment members 91a and 91b receive the coupled channel signal from the channel 27 and receive the scale factor from the scale factor calculator 44, and perform processing equivalent to that achieved in the decoding process, which will be discussed below. To generate a decoupling signal from the coupled 5-channel signal. The decoupling signal is transmitted to the synthesis modes 2la* and 21b, and the scale factor is calculated in a manner similar to that discussed above with respect to FIG. · In the additional fabrication of the encoder shown in Figure 6, if the frequency component is not required to calculate the spectral energy measurement and scaling factor, then the composite modes Lu 10 21a and 21b may be independent of the baseband signal and/or Operated by the spectral components of the coupled channel signal. In addition, if HFR* uses the spectral components in the Cannes channel signal, the synthesis mode can be made regardless of the channel signal being tapped. 2. Decoder 15 20 Figure 7 shows a group audio decoder that receives two channels from channel S9.

The group of the singer's singer is coded and truncated along the channel and 8% produces a decoded representation of the nickname. The formatter 6〇, imitation (4), signal scale components 70a and ., are::& components are not the details and features of the single frequency ship are substantially the same as those of the above; the pool pool and the channel decoder. The deformatter 60 derives the four components from the encoded signal and the set of scale factors. The signal is lightly combined, and the input signal is input into the audio signal towel. The composite material represents the two sets of the two sets of input audio signals. The scale factors of each of the two sets of input audio signals are respectively transmitted along the channels 63a and 63b. Signal scale component 92a produces a spectral component of a set of decoupled signals along channel 93a that is close to the spectral energy level of the corresponding spectral component of a set of original input audio signals. These decoupled spectral components can be generated by multiplying the spectral components of the combined 5-channel signal with an appropriate coupling scaling factor. In the process of configuring the spectral component of the coupled channel signal in the frequency sub-band and providing a set of scale factors for each sub-band, the spectral components of the decoupled signal can be generated according to the expression of (11): XD.Xk) = SF-Xm ) - XC{k) (11) 10 where XC(k) = the kth group of spectral components in the mth sub-band of the coupled channel signal; SFi(m) = the mth frequency of the i-th signal channel The scale factor of the frequency band; and XDi(k) = the kth group of decoupled spectral components of the i-th signal channel. 15 Each decoupled signal is passed to a separate set of synthesis filter banks. In the above preferred production, the spectral components of the de-combined signals are in one or more sub-bands of the third group of frequency sub-bands in the middle of the frequency sub-bands of the _th group and the second group of frequency sub-bands. . If they require signal synthesis, then the decoupled spectral components are simultaneously transmitted 20 to a respective set of signal synthesizing members 23a or 23b. E. Adaptive Aggregation The coding system in which the spectral components are configured as two or three sets of frequency sub-bands as discussed above can be adapted to the frequency range or sub-band range included in each group. For example, when having an input audio signal interval that is considered to be similar to the component of the high frequency spectrum 31 1324762 of noise, it is advantageous to reduce the lower end portion of the second set of frequency subband frequency ranges of the residual signal. The frequency range can also be adapted at the same time to remove all sub-bands in a set of frequency sub-bands. For example, HFR 5 processing of an input audio signal having a large, abrupt amplitude change can be disabled by removing all sub-bands from the second set of frequency sub-bands. Figures 3 and 4 illustrate a method in which the frequency range of the baseband, residual, and/or coupled channel signals can be adapted for any reason including one or more sets of characteristics that are reflected to the input audio signal. To perform this feature, each of the analysis filter banks shown in Figures 1, 5, 6 and 8 can be replaced by the device shown in Figure 3 and the synthesis filter banks shown in Figures 2 and 7 can be viewed in Figure 4. Replaced by the display device. These graphical display frequency sub-bands can be adapted for three sets of frequency sub-bands; however, the same fabrication principles can be used to accommodate different sets of sub-bands. Referring to Figure 3, the analysis filter bank 14 receives a set of input audio signals from the channel 9 and reactively produces a set of frequency sub-band signals that are transmitted to the adaptive focusing member 15. Signal analysis component 17 analyzes the information derived directly from the input audio signal and/or derived from the sub-band signal and reacts to the analysis to produce band control information. The band control information is transmitted to the adaptive aggregation component 15 and it transmits band control information to the formatter 50 along the channel 18. The formatter 50 contains a set of representations of this band control information in the encoded signal. The adaptive aggregation component 15 reacts to the band control information using the specified sub-band signal spectral components to the frequency sub-band set. The spectral components assigned to the first set of 32 bands are transmitted along channel 12. The spectral components assigned to the second set of secondary bands are transmitted along channel 11. The spectral components assigned to the third set of sub-bands are transmitted along channel 13. If there are frequency ranges or gaps that are not included in any of the collections, this can be achieved by not arranging the spectral components in the range or gap to any set. The signal analysis component 17 can also generate band control information to adjust the frequency range in response to conditions that are unrelated to the input audio signal. For example, the range can be adapted in response to a signal representative of the level required to transmit or record the available capacity or signal quality of the encoded signal. Band control information can be generated in many forms. In a production, the band control information specifies the lowest and/or highest frequency for each group to which the spectral components will be assigned. In another production, the band control information specifies a set of frequency ranges for a plurality of predetermined configurations. Referring to Fig. 4, the adaptive aggregation member 81 receives a set of spectral components from the channels 71, 93 and 62 and receives band control information from the channel 68. The formatter 60' band control information is derived from the encoded signal. The adaptive component set 81 enters a set of frequency sub-band signals by distributing the spectral components in the set of spectral components received, which are passed to the synthesis filter bank 82 for reaction to the band control information. The synthesis filter bank 82 reacts to a frequency of one person frequency ▼ 彳 5 and generates a set of output audio signals along channel 89. F. The second analysis filter bank performs a spectral energy measurement of the expression la, such as an employee-to-input input, in the audio encoder of the analysis chopper library using a conversion (eg, TDAC conversion described above). The exact frequency energy of the audio signal is low, since the 1324462 provides only real-valued conversion coefficients for the analysis filter bank. Fabrication using the same transformation as discrete Fourier transform (DFT) can provide a more accurate energy meter calculation because each conversion coefficient is represented by a complex value that more accurately conveys the true amplitude of each spectral component. 5 According to the inherent inaccuracy of the conversion coefficient energy calculation using only real values converted from, for example, TDAC, a second analysis filter having a basis function orthogonal to the basis function of the analysis 鬌 filter bank 10 can be used. The library was overcome. Figure 8 shows a set of audio encoders that are similar to the encoder shown in Figure 1, but which includes a second set of analysis filter banks 19. If the X10 encoder uses the TDAC converted MDCT to perform the analysis filter bank 10, a corresponding set of modified discrete sinusoidal transforms (M D S T) can be used to create the second analysis filter bank 19. The energy calculator 39 calculates a more accurate spectral energy measurement E'(k) from the expression (12): 15 E'{k) = Xl2{k) + X22{k) (12) where XJkk is from the first analysis filter The kth group of conversion coefficients of the bank; and _X2(k) = the kth group of conversion coefficients from the second analysis filter bank. In the production of the energy measurement of the calculation frequency sub-band, the energy calculator 39 calculates the measurement of the m-th frequency sub-band from the expression (13): , 20 E'(m) = VZ, 2(A:) + Z22 (A:) (13) The scale factor calculator 49 calculates the scale factor W(m) from these more accurate energy measurements in a manner similar to that of Expression 3a or 3b. A calculation similar to the expression 3a is shown in Expression 14. η 14. 34 ___ (14) SF'(m) = f£@V (w) mtf}

Care should be taken when using the scale factor (m) calculated from these more accurate energy measurements. The spectral division of the synthesized signal, which is scaled according to the more precise scale factor w((7)), will almost certainly distort the relative spectral balance of the 5 part of the signal baseband and the synthesized portion of the regeneration, because of the more accurate energy measurement It will always be greater than or equal to the energy measurement from the real number conversion factor. One way in which this difference can be compensated is to reduce the more accurate energy measurement by half, because, on average, more accurate measurements will be twice as large as the less accurate measurements. This reduction will provide the advantage of maintaining a more accurate spectral energy measurement in the signal baseband and a statistically consistent energy level in the combined portion. The denominator of the ratio number in Expression 14 should be calculated only from the real conversion coefficients from the analysis filter bank 10, even if additional coefficients from the second analysis filter bank 19 are available. The calculation of the scale factor should be

15 Completion' because during the decoding process, the scale adjustment will be based on the synthesized spectral components of the conversion coefficients obtained only similar to the self-analysis filter bank 10. The decoding process will not access any coefficients that correspond to or can be derived from the spectral components from the second analysis filter bank 19. G. Production 20 Various aspects of the present invention can be implemented in a variety of ways, including in general purpose computer systems or software in some other devices, including more specialized components such as digital signal processors (DSPs). The circuit, 35 1324762 is transmitted by storing media, for example, in a baseband or a modulated communication channel including a spectrum from ultrasonic waves to ultraviolet frequencies, and the storage medium substantially uses a magnetic tape, a card or a disc, and an optical card. Or any recording technique for discernable marks on optical discs, and paper-like media, to transmit information. BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is an exploded block diagram of an apparatus for encoding an audio signal for subsequent decoding by means of a high frequency reproduction device. Figure 2 is an exploded block diagram of a device for decoding an encoded audio signal using high frequency reproduction. Figure 3 is an exploded block diagram of a device for dividing an audio signal into a frequency sub-band signal having a range adapted to reflect one or more of the characteristics of the audio signal. Figure 4 is an exploded block diagram of an apparatus for synthesizing 15 audio signals from frequency sub-band signals having an adapted range. Figures 5 and 6 are exploded block diagrams of the apparatus for encoding an audio signal for coupling for subsequent decoding by means of high frequency reproduction and decoupling. Figure 7 is an exploded block diagram of a device that uses high frequency reproduction and decoupling to decode the encoded audio 20 signal. Figure 8 is an exploded block diagram of an apparatus for encoding an audio signal using a second set of analysis filter banks to provide additional spectral components for energy calculation. Figure 9 is an exploded block diagram of an apparatus in which various aspects of the present invention can be made. 37 1324762 [The main components of the diagram represent the symbol table] 9...channel 9a...channel 9b...channel 10...analysis filter bank 10a...analysis filter library 1 Ob...analysis filter library 11...channel 11a...channel lib... Channel 12...channel 12a...channel 12b...channel 13a...channel 13b...channel 14...analysis of the waver library 15...adaptive aggregation member 16...channel 17...signal analysis component 18...channel 19...second analysis of the waver library 21...synthesis Mode 21a···Synthesis mode 21b...Synthesis mode 23...Signal synthesis component 23a···Signal synthesis component 23b...Signal synthesis component 26···Coupler 27...Channel 31...Energy calculator 31a···Energy calculator 31b ··· Energy calculator 32...Energy calculator 32a···Energy calculator 32b...Energy calculator 35...Energy calculator 35a···Energy calculator 35b···Energy calculator 37...Energy calculator 39...Energy Calculator 40...Scale Factor Calculator 40a...Scale Factor Calculator 40b...Scale Factor Calculator 41...Channel 41a...Channel 41b...Channel 44...Scale Factor Calculator

38 1324762 45a...channel 71b...channel 45b...channel 72 ...DSP 49...scale factor calculator 73--RAM 50...formatter 74 ...ROM 51...channel 75...I/O control 59...channel 76...communication channel 60...solution Formatter 77...communication channel 61...channel 80...synthesis filter bank 61a...channel 80a...synthesis filter bank 61b...channel 80b...synthesis filter bank 62...channel 81...adaptive aggregation member 62a...channel 82...synthesis filter Library 62b...channel 89...channel 63...channel 89a...channel 63a...channel 89b...channel 63b...channel 91a...scale adjustment member 64...channel 91b...scale adjustment member 68...channel 92a...signal scale member 70...signal scale member 92b... Signal Scale Member 70a...Signal Scale Member 93...Channel 70b...Signal Scale Member 93a...Channel 71...Channel 71a...Channel 93b...Channel

39

Claims (1)

1324762 Picking up, patenting scope: 1. A method for encoding one or more sets of input audio signals, wherein the method comprises: receiving the one or more sets of input audio signals and obtaining a 5 or more sets therefrom a baseband signal and one or more sets of residual signals, wherein a spectral component of the set of baseband signals represents a spectral component of a respective set of input audio signals in the first set of frequency subbands and a spectral component of the associated residual signal represents a spectral component of the input audio signal in a second set of frequency sub-bands not represented by the baseband signal; 10 obtaining an energy measure of at least some of the spectral components that are generated in one or more combined signals during decoding And wherein the one or more combined signals have spectral components in a second set of frequency sub-bands; an energy measurement of at least some spectral components of each residual signal is obtained; and a scaling factor is calculated, which is utilized to obtain spectral component energy in the residual signal 15 Measure the square root of the ratio of the energy measurement of one or more of the signals to the IF reading component, in one or more The ratio of the square root of the ratio of the energy content of the spectral component in the composite signal to the energy measurement of the spectral component in the residual signal, the square root of the energy component of the spectral component of the residual signal, and the square root of the energy content of the spectral component in one or more of the combined signals. Calculate the scale factor by calculating the ratio of the square root of the spectral component energy measurement in one or more combined signals to the square root of the spectral component energy measurement in the residual signal; and combining the signal information and the scale information into a set of codes A signal, wherein the signal information represents a frequency component of the frequency 40 1324762 in the one or more sets of baseband signals and the scale information represents the scale factor. 2. The method of claim 1, wherein the one or more combined signals are generated at least in part by frequency conversion of at least some of the spectral components of the one or more sets of baseband signals. 5. The method of claim 2, wherein the spectral component of the synthesized signal is generated using frequency conversion that maintains phase coherence. 4. The method of claim 1, wherein the one or more combined signals are generated, at least in part, utilizing frequency conversion of at least some of the spectral components of the one or more sets of baseband signals and having spectral bits Quasi 10 is generated by adapting a combination of one or more sets of noise-like signals based on a spectral level in the one or more sets of baseband signals, and wherein the one or more combined signals are generated Spectral component energy measurements are obtained without regard to the spectral level in the noise-like signal. 5. The method of claim 1, wherein the one or more combinations of signals are generated at least in part by the generation of one or more sets of noise-like signals. 6. The method of claim 1, wherein the spectral component energy measurement of the residual signal is obtained from a value representative of the amplitude of the spectral component. 20. The method of claim 6, comprising: applying a set of first analysis chopper banks to the one or more sets of input audio signals to obtain the one or more sets of baseband signals and the one One or more sets of residual signals; and applying a set of second analysis filter banks to the one or more sets of inputs 41 1324762 for measurement; obtaining two sets of signals represented by the coupled channel signals in the third set of frequency sub-bands Or more sets of energy measurements of at least some of the spectral components of the input audio signal; and 5 calculating a coupling scale factor that utilizes the spectral component energy measurements in the two or more sets of input audio signals for the coupled channel signal The square root of the ratio of the energy measurement of the spectral energy, the square root of the ratio of the spectral energy energy measurements in the coupled channel signal to the spectral component energy measurements in the two or more sets of input audio signals, the two 10 groups or Ratio of the square root of the spectral component energy measurement in the set of input audio signals to the square root of the spectral energy energy measurement in the coupled channel signal Or calculating a coupling scale factor for the ratio of the square root of the spectral energy energy measurement in the coupled channel signal to the square root of the spectral component energy measurement in the two or more sets of input audio signals; 15 wherein the scale information also represents The coupling scale factor and the signal information also represent the spectral components in the coupled channel signal. 15. The method of claim 14, wherein the one or more combined signals are at least partially utilized by frequency conversion of at least 20 of the spectral components of the input audio signals in the third set of frequency sub-bands Was produced. 16. The method of claim 14, comprising: detecting one or more sets of characteristics of the plurality of input audio signals; adapting to the detected characteristic, adapting the first set of frequency sub-bands, 43 1324762 second group Frequency sub-band, or third-rate sub-band frequency and solid, combining a set of adapted frequency range indications according to the method of claim 17. According to the method of claim 1, which includes: 5 detecting the One or more sets of one or more sets of input audio signals; and, responsive to the detected characteristic, adapting the frequency range of the second set of frequency sub-bands of the first set of frequency sub-bands; and combining a set of The indication of the adaptation frequency range is for the encoded signal. 10 I8· A method for decoding an encoded signal representing one or more sets of input audio signals, wherein the method comprises: obtaining scale information and signal information from the encoded signal, wherein the scale information represents an amount of energy by a spectral component The square root of the ratio or the square root ratio of the energy component of the frequency component is calculated as a scale factor, and the 15 signal information represents a spectral component of one or more sets of baseband signals, wherein the spectral components in each of the baseband signals represent the first a spectral component of the input audio signal in a set of frequency sub-bands; generating a set of correlated synthesized signals for each of the respective baseband signals, the composite signal having a second set of frequencies not represented by the respective baseband signals a spectral component in the sub-band that causes the spectral components of the correlated composite signal to be scaled by multiplication or division according to one or more sets of scale factors; and one or more sets of output audio signals are generated The output audio signal represents a set of separate input audio signals and is derived from a respective fundamental frequency 44 The component of the band signal and its associated synthesized signal is generated. 19. The method of claim 18, wherein the correlated synthesized signal is generated at least in part by frequency conversion of at least some of the spectral components of the respective baseband signals. 5 20. The method according to claim _, wherein the frequency conversion maintains phase coherence. 21. According to the method of the patent scope (4), the relevant information is synthesized. |5 Injury is generated using the generation of noise-like signals with spectral levels that are adapted according to one or more sets of scale factors. H) 22. According to the method of claim 18 of the patent, it obtains one or more sets of normalized values from the encoded signal and normalizes the inverse scale factor with respect to the set or sets of normalized values. . 23. The method according to claim 22, wherein the one or more sets of positive values are transmitted to the encoded signal using a scale 15 information of the selected value in the set of representative values. 24. The method according to claim 22, wherein the one or more sets of normalized values comprise a scale factor - a maximum allowable value. 25. The method according to claim 18, wherein the frequency sub-band of the correlated synthesized signal is related to a respective scale factor of the group. 2. According to the method of claim 25, it is adapted to the generation of the associated synthesized signal in response to the sub-band information transmitted in the encoded signal specifying the frequency range of the frequency sub-band. 27. The method of claim 26, wherein the sub-band information represents a selected frequency range in the set of ranges. The 45 1324762 signal is also generated from the spectral components of the respective decoupled signals. 29. The method of claim 28, wherein the correlated synthesized signal is generated at least in part by frequency conversion of at least some of the spectral components in the third set of frequency sub-bands. 5: The method of claim 28, comprising: obtaining a frequency range indication of the first, second or third set of frequency sub-bands from the encoded signal; and adapting the indication to the synthesized signal and The decoupling signal is generated. 10: The method according to claim 18, comprising: obtaining a frequency range indication of the first or second set of frequency sub-bands from the encoded signal; and adapting the synthesized signal and being decoupled in response to the indication The generation of signals. 15 32, a method for encoding a plurality of input audio signals, wherein the method comprises: receiving a plurality of input audio signals and obtaining a plurality of baseband signals, a plurality of residual signals, and a set of coupled channel signals therefrom, The spectral components of the set of baseband signals represent spectral components of the input audio signals of one of the first set of frequency sub-frequency 20 bands and the spectral components of a correlated residual signal represent the first not represented by the baseband signal a spectral component of the input audio signal of the two sets of frequency sub-bands, and wherein the spectral component of the coupled channel signal represents one of 47 1324762 of the two or more sets of input audio signals in the third set of frequency sub-bands Combining; obtaining respective residual signals and energy measurements of at least some of the spectral components of the two or more sets of input audio signals represented by the coupled channel signals; and 5 combining the control information and signal information with a set of encoded signals, Where the control information is derived from the energy measurement and wherein the signal information represents a majority The baseband signal and the spectral component of the coupled channel signal. 33. The method of claim 32, comprising: 10 obtaining an energy measurement of at least some of the spectral components that are generated as one or more combined into a signal at the time of decoding, wherein the one or more combined signals have The spectral components in the second set of frequency sub-bands; and at least some control information is derived by calculating a square root of the ratio of the energy measurements or a square root ratio of the energy measurements. 15 34. The method of claim 33, wherein at least some of the spectral components of the one or more combined signals are synthesized from spectral components of the third set of frequency subbands. 35. The method of claim 32, wherein the frequency range of the frequency sub-band set is adapted, and wherein the method combines the indication of one of the adapted frequency ranges with the encoded signal. 36. A method for decoding an encoded signal representative of a plurality of input audio signals, wherein the method comprises: obtaining control information and signal information from the encoded signal, wherein the control information is an energy measurement from a spectral component Derived and the signal 48 1324762 information represents a spectral component of a plurality of baseband signals and a set of coupled channel signals, wherein the spectral components in each of the baseband signals represent a respective set of input audio signals of the first set of frequency subbands a spectral component and the spectral component of the coupled channel signal represents a composite of a spectral component of a third set of frequency sub-bands of two or more sets of the plurality of input 5 audio signals; generating for each respective baseband signal a set of correlated synthesized signals having spectral components in a second set of frequency sub-bands not represented by the respective baseband signals, wherein the spectral components of the correlated synthesized signal 10 are adjusted in accordance with the control information a scale generated from the coupled channel signal for use with the coupled channel signal a respective set of decoupling signals of each of the groups of input audio signals, wherein the decoupled signal has 15 spectra in the third set of frequency sub-bands scaled according to the control information; Generating a plurality of output audio signals, wherein each of the output audio signals represents a respective set of input audio signals and is generated from a set of respective baseband signals and spectral components of their associated synthesized signals, and wherein the two sets or The output audio signals of more group audio signals are also generated from the spectral components of the respective decoupled signals. 37. The method according to claim 36, wherein the control information conveys a set of scale factors calculated from a square root of the energy measurement ratio or a square root ratio of the energy measurements, and wherein some of the ratios The energy measurements represent at least some of the spectrum of the synthesized signals as 49 1324762 parts of energy. 38. The method of claim 37, wherein at least some of the spectral components of the one or more combined signals are synthesized from spectral components in the third set of frequency sub-bands. 5 39. The method according to claim 36, wherein one or more sets of frequency ranges of the frequency sub-band are adapted in response to the control information. 40. An encoder for encoding one or more sets of input audio signals, wherein the encoder has processing circuitry for performing a signal processing method, the method comprising: 10 receiving the one or more sets of input audio signals and Wherein one or more sets of baseband signals and one or more sets of residual signals are obtained, wherein the spectral components of the set of baseband signals represent spectral components of a respective set of input audio signals in the first set of frequency subbands and are correlated The spectral components in the residual signal represent the spectral components of the input audio signals in the 15th frequency sub-bands not represented by the baseband signal; the resulting one or more combined signals are generated during decoding. An energy measurement of at least some of the spectral components, wherein the one or more combined signals have spectral components in a second set of frequency sub-bands; an energy measurement of at least some spectral components of each residual signal is obtained; 20 calculating a scaling factor, It utilizes the ratio of the energy content of the spectral components in the residual signal to the energy measurement of one or more of the components of the signal. Square root, the square root of the ratio of the energy content of the spectral components in one or more of the combined signals to the energy measurements of the spectral components in the residual signal, and the square root of the energy measurement spectral components in the residual signal for a group of 50 or more combined signals The ratio of the square root of the energy component of the spectral component, or the ratio of the square root of the energy component of the spectral component in one or more of the combined signals to the square root of the spectral component energy measurement in the residual signal; and the group ek number The information and scale information becomes a set of encoded signals, wherein the signal information represents spectral components in the one or more sets of baseband signals and the scale information represents the scale factor. A decoder for decoding an encoded signal representing one or more sets of input audio signals, wherein the decoder has a processing circuit for performing a signal processing method, the method comprising: obtaining scale information and signals from the encoded signal Information, wherein the scale machine represents a scale factor calculated from a square root of a spectral component energy measurement ratio or a square root ratio of a spectral component energy measurement, and the signal information represents a spectral component of one or more sets of baseband signals, wherein The spectral components in each of the baseband signals represent spectral components of the respective input audio signals in the first set of frequency subbands; a set of correlated synthesized signals are generated for each of the respective baseband signals, the synthesized signals having the respective a spectral component of a second set of frequency sub-bands represented by the baseband signal, wherein the spectral components of the correlated composite signal are scaled by multiplication or division according to one or more sets of scale factors; and a set of or Multiple sets of output audio signals, wherein each output audio signal represents a separate set of And the audio signal from a respective base band signal and the correlation signal is synthesized spectral components are generated. 1324762 42. An encoder for encoding a plurality of input audio signals, wherein the encoder has a processing circuit for performing a signal processing method, the method comprising: receiving a plurality of input audio signals and obtaining a plurality of bases therefrom a frequency band signal, a plurality of residual signals, and a set of coupled channel signals, wherein a spectral component of the set of baseband signals represents a spectral component of the input audio signal of one of the first set of frequency subbands and an associated residual signal The spectral components represent spectral 10 components of the respective input audio signals in the second set of frequency sub-bands not represented by the baseband signal, and wherein the spectral components of the coupled channel signals represent two of the third set of frequency sub-bands or Combining one of a plurality of spectral components of the input audio signal; obtaining each residual signal and energy measurement of at least some of the spectral components of the two or more sets of input audio signals represented by the coupled channel signal; and combining Controlling information and signal information in a set of encoded signals, wherein the control information is from the Measuring the amount of the signal is derived, and wherein the information represents a plurality of base band signal and the channel signals are coupled to parts of the spectrum. 20 43. A decoder for decoding an encoded signal representing a plurality of input audio signals, wherein the decoder has a processing circuit for performing a signal processing method, the method comprising: obtaining control information from the encoded signal and Signal information, wherein the control information is derived from energy measurements of spectral components and the signal 52 1324762 information represents spectral components of a plurality of baseband signals and a set of coupled channel signals, wherein spectral components in each of the baseband signals represent a spectral component of the input audio signal of one of the first set of frequency subbands and the spectral component of the coupled channel signal representing a third set of frequency subbands of two or more sets of the plurality of input 5 audio signals Generating a set of spectral components; generating a set of correlated synthesized signals for each of the respective baseband signals, the synthesized signals having spectral components in a second set of frequency subbands not represented by the respective baseband signals, wherein the correlation The spectral components in the synthesized signal 10 are scaled according to the control information; The coupled channel signal produces a respective set of decoupling signals for each of the two or more sets of input audio signals represented by the coupled channel signal, wherein the decoupled signal has a scale that is adjusted according to the control information The spectrum in the third set of frequency sub-bands is 15; and a plurality of output audio signals are generated, wherein each output audio signal represents a set of separate input audio signals and from a respective set of baseband signals and their associated synthesized signals The spectral components are generated, and the output audio signals representing the two or more sets of audio signals are also generated from the spectral components of the respective decoupled signals. 44. A medium for transmitting a program of instructions executable by a device, wherein execution of the instructions causes the apparatus to perform a method for encoding one or more sets of input audio signals, wherein the method comprises: receiving the set Or a plurality of sets of input audio signals and from which a 53 1324762 group or groups of baseband signals and one or more sets of residual signals are obtained, wherein a spectral component of a set of baseband signals represents a group of the first set of frequency subbands a spectral component of the input audio signal and a spectral component of a correlated residual signal representing a spectral component of the input audio signal in a second set of frequency subbands not represented by the baseband signal; obtained at the time of decoding One or more of the energy measurements of at least some of the spectral components of the signal, wherein the one or more combined signals have spectral components in the second set of frequency sub-bands; Obtaining an energy measurement of at least some of the spectral components of each residual signal; calculating a scaling factor that utilizes the square root of the ratio of the spectral component energy measurements in the residual signal to one or more of the spectral component energy measurements in the signal, The square root of the ratio of the energy content of the spectral components in the signal or the combined signal, the square root of the ratio of the energy components of the residual signal, and the energy content of the spectral components in the residual signal. The ratio of the square root, or the ratio of the square root of the spectral component energy measurement in one or more of the combined signals to the square root of the spectral component energy measurement in the residual signal; and the combined signal information and scale information becomes a group An encoded signal, wherein the signal information represents a spectral component in the one or more sets of baseband signals and the scale information represents the scale factor. 4S•~ a medium for transmitting a program program executable by a device, wherein execution of the program causes the device to perform a method for decoding an encoded signal representing a _ level or groups of input audio signals, wherein The method of reference 54 1324762 includes: obtaining scale information and signal information from the encoded signal, wherein the scale information represents a scale factor calculated from a square root of a spectral component energy measurement ratio or a square root ratio of a spectral component energy measurement, and the scale factor is 5 The signal information represents a spectral component of one or more sets of baseband signals, wherein the spectral components in each of the baseband signals represent spectral components of respective input audio signals in the first set of frequency subbands; and are generated for respective respective baseband signals a set of correlated synthesized signals 'the synthesized signal having a spectral component in a 10th set of frequency sub-bands not represented by the respective baseband signals' wherein the spectral components of the correlated synthesized signal are based on one or more sets of scales Factor is scaled by multiplication or division; and produces one or more Output audio signal, wherein each of the output audio signals represent a set of audio input signals, respectively, and 15 respectively with signals from a baseband signal, and the correlation was synthesized spectral components are generated. 46. The medium according to claim 45, wherein the associated synthesized signal is generated at least in part by frequency conversion of at least some of the spectral components of the respective baseband signals. 47. The medium according to claim 46, wherein the frequency conversion maintains 20 phase coherence. 48. The medium according to claim 45, wherein the associated synthesized signal is generated, at least in part, by the generation of a noise-like signal having a spectral level adapted according to one or more sets of scale factors. 49. According to the medium of claim 45, the encoded signal obtains the square root of the ratio of the spectral component energy measurements of the two or more sets of input audio signals from the encoded signal, in the third set of frequency subbands The ratio of the square root of the spectral component energy measurement of the two or more sets of input audio signals to the square root of the spectral energy energy measurement in the coupled channel signal, or the amount of the spectral energy energy in the coupled channel signal Calculating a square root for the ratio of the square root of the spectral component energy measurements of the two or more sets of input audio signals in the third set of frequency sub-bands; and generating the utilized coupled channel signal 10 from the coupled channel signal Decoding signals respectively for each of the two or more sets of input audio signals, wherein the de-combined signals are scaled by multiplication or division according to one or more sets of scale factors a spectral component of the third set of frequency sub-bands; wherein the output audio signal representing the two or more sets of input audio signals Is decoupled from the signals generated in the spectral component. 56. The medium according to claim 55, wherein the correlated synthesized signal is generated at least in part by frequency conversion of at least some of the spectral components in the third set of frequency sub-bands. 57. The medium according to claim 55, wherein the method comprises: 20 obtaining a frequency range indication of the first, second or third set of frequency sub-bands from the encoded signal; and adapting to the indication and adapting to be synthesized The generation of signals and decoupled signals. 58. The medium according to claim 45, wherein the method comprises: 57 1324762 obtaining a frequency range indication of the first or second set of frequency sub-bands from the encoded signal; and adapting the synthesized signal to the indication The decoupling signal is generated. 5 59. A medium for transmitting a program program executable by a device, wherein execution of the program causes the device to perform a method for encoding a plurality of input audio signals, wherein the method comprises: receiving a plurality of input audio signals And obtaining a plurality of baseband signals, a plurality of residual signals, and a set of coupled channel signals, wherein a spectral component of one of the baseband signals represents a set of input audio signals of the first group of frequency subbands The spectral components and a spectral component of the associated residual signal represent spectral components of the respective input audio signals in the second set of frequency sub-bands not represented by the baseband signal, and wherein the spectral components of the coupled channel signals represent the third component Combining one of two or more sets of spectral components of the input audio signal in the 15 frequency subbands; obtaining at least some spectral components of each residual signal and two or more sets of input audio signals represented by the coupled channel signal Energy measurement; and 20 combined control information and signal information are grouped Signal, wherein the control information is derived from the measured amount of energy, and wherein the information signal represents a plurality of base band signal and the channel signals are coupled to parts of the spectrum. 60. A medium for transmitting a program program executable by a device, wherein execution of the 58 1324762 instruction program causes the device to perform a method for decoding a group of encoded signals representative of a plurality of input audio signals, wherein the method The method includes: obtaining control information and signal information from the encoded signal, wherein the control information is derived from energy measurement of a spectral component and the signal information represents a spectral component of a plurality of baseband signals and a set of coupled channel signals a spectral component of each of the baseband signals representing a spectral component of the input audio signal of one of the first set of frequency subbands and the spectral component of the coupled channel signal representing two of the plurality of input 10 audio signals or A plurality of spectral components of the third set of frequency sub-bands of the further group are composited; a set of correlated synthesized signals are generated for each of the respective baseband signals, the composite signal having a second set of signals not represented by the respective baseband signals a spectral component in a frequency sub-band, wherein the correlation is a spectral component of the synthesized signal 15 Adjusting a scale according to the control information; generating, from the coupled channel signal, a set of respective decoupling signals for each of two or more sets of input audio signals represented by the coupled channel signal, wherein the decoupling The signal has 20 spectra in the third set of frequency sub-bands that are scaled according to the control information; and generates a plurality of output audio signals, wherein each output audio signal represents a set of separate input audio signals and is from a group Spectral components of the respective baseband signals and their associated synthesized signals are generated, and wherein the output audio signals representing the two or more sets of audio signals are also from the frequency components of the respective decoupled signals of 59 1324762 produce. 61. The medium according to claim 60, wherein the control information conveys a set of scale factors calculated from a square root of the energy measurement ratio or a square root ratio of the energy measurements, and wherein 5 of the ratios Some energy measurements represent the energy of at least some of the spectral components of the synthesized signals. 62. The medium of claim 61, wherein at least some of the spectral components of the one or more combined signals are synthesized from spectral components in the third set of frequency subbands. 10 63. The medium according to claim 60, wherein one or more sets of frequency ranges of the frequency sub-band are adapted in response to the control information. 64. A medium for transmitting encoded information representative of one or more sets of input audio signals, wherein the encoded information is generated using a method, the method comprising: 15 receiving the one or more sets of input audio signals and from the Obtaining one or more sets of baseband signals and one or more sets of residual signals, wherein a spectral component of the set of baseband signals represents a spectral component of a respective set of input audio signals in the first set of frequency subbands and an associated The spectral components in the residual signal represent spectral components of the input audio signals in the 20th second frequency sub-band not represented by the baseband signal; and at least one or more combined signals are generated at the time of decoding. An energy measurement of a plurality of spectral components, wherein the one or more combined signals have spectral components in a second set of frequency sub-bands; an energy measurement of at least some spectral components of each residual signal is obtained; 60 1324762 spectral energy in the signal The square root of the ratio of the energy measurements, the spectral energy energy measurement in the coupled channel signal for the two or more groups The square root of the ratio of the spectral component energy measurements in the input audio signal, the square root of the spectral component energy measurements in the two or more sets of input audio signals for the spectral energy energy measurements in the coupled channel signal The ratio of the square root, or the square root of the spectral energy energy measurement in the coupled channel signal, calculates a coupling scale factor for the ratio of the square root of the spectral component energy measurements in the two or more sets of input audio signals; wherein the scale information is simultaneously Also representing the coupling scale factor and the 10 signal information also represents the spectral component of the coupled channel signal. 66. The medium according to claim 65, wherein the one or more combined signals are at least partially utilized by frequency conversion of spectral components of at least some of the input audio signals in the third set of frequency sub-bands Produced 15 students. 67. The medium according to claim 65, wherein the method comprises: detecting one or more sets of characteristics of a plurality of input audio signals; adapting to the detected characteristic, adapting the first set of frequency sub-bands, the second set a frequency sub-band, or a frequency range of a third set of frequency sub-bands; 20 and combining a set of adapted frequency ranges indicative of the encoded signal. 68. The medium according to claim 64, wherein the method comprises: detecting one or more sets of characteristics of the one or more sets of input audio signals; 62 1324762 reacting to the detected characteristic and adapting the first set of frequencies A frequency range of the sub-band or the second set of frequency sub-bands; and combining a set of adapted frequency ranges indicative of the encoded signal. 69. A medium 5 for transmitting encoded information representative of a plurality of input audio signals, wherein the encoded information is generated by a method, the method comprising: receiving a plurality of input audio signals and obtaining a plurality of basebands therefrom a signal, a plurality of residual signals, and a set of coupled channel signals, wherein a spectral component of the set of baseband signals represents a spectral component of the input audio signal of one of the first set of frequency subbands and a residual signal of one phase 10 The spectral components represent spectral components of the respective input audio signals in the second set of frequency sub-bands not represented by the baseband signal, and wherein the spectral components of the coupled channel signals represent two of the third set of frequency sub-bands or Combining one of a plurality of spectral components of the input audio signal; 15 obtaining an energy measurement of each residual signal and at least some spectral components of the two or more sets of input audio signals represented by the coupled channel signal; and combining control Information and signal information in a set of encoded signals, wherein the control information is measured from the energy And wherein the derived signal information table 20 on behalf of a plurality of base band signal and the channel signals are coupled to parts of the spectrum. 70. The medium according to claim 69, wherein the method comprises: obtaining an energy measurement of at least some spectral components that are generated one or more combined into a signal at the time of decoding, wherein the one or more combinations are 63 1324762 having spectral components in the second set of frequency sub-bands and at least some of the spectral components of the one or more combined signals are synthesized from the spectral components of the third set of frequency sub-bands t; and utilizing the calculated amount of energy At least some control information is derived by measuring the square root of the ratio or the square root ratio of the energy measurement. 71. The medium of claim 69, wherein the frequency range of the frequency sub-band set is adapted, and wherein the method combines the indication of one of the adapted frequency ranges with the encoded signal.